20 BULLETIN 796, U. S. DEPARTMENT OE AGRICULTURE. 
The trees treated were about 10 years of age and had suffered 
heavily from borer attacks. The} 7 received 1 ounce each. On May 
28, 1917, 15 were examined. Nine showed no injury at all, 5 showed 
traces, and 2 what was classified as slight injury. So far as the final 
effect on the trees was concerned none of this injury was of impor- 
tance. It was more than hitherto had been observed for this dose, 
however, and while the reason is not entirely apparent it is felt 
that it is best to make the application sufficiently early in the fall 
to allow for complete evaporation before winter. This fact, coupled 
with its less effective action, should be sufficient reason for avoiding 
late application. When the application is made in the central lati- 
tude as late as September 15, it probably would be wise to uncover 
the trees after five or six weeks, and either allow them to stand open 
for a time or refill with fresh earth. In fact, at the present stage of 
our knowledge of the problem the writer is inclined to feel that this 
might be a wise precaution to follow in every case. 
The varieties treated have included many of the leading commer- 
cial sorts as well as a large number of unknown seedlings. So far 
there has been nothing to indicate that one variety of peach is more 
susceptible to the effects of the gas than another. 
A summary of the results obtained by one fall application of para- 
dichlorobenzene is given in Table VI. 
RELATION OF INJURY TO SOIL TYPE. 
Xo special relation between injury and type of soil has been discov- 
ered in the use of p-dichlorobenzene. Very porous soils probably 
give a somewhat less concentrated vapor than very retentive soils, 
. but in practice the effect of soil type has not seemed important, al- 
though the soils on which experiments were carried on have varied 
from light sandy loams to heavy clay loams. 
INJURY TO APPLE. 
In August, 1916, an application of p-dichlorobenzene was made to 
twelve 3-year-old apple trees at Springfield, W. Va. These trees were 
treated with doses varying from 1 ounce to one-fourth of an ounce 
each. All were quite well infested with S. Candida. The applica- 
tion was made on August 23 and the examination on September 1-2. 
Practically all the smaller larvae of the borer had been killed. The 
more mature specimens working in deeper burrows were still living, 
although several were affected noticeably by the gas. The trees were 
severely injured, however. Apparently, the action of the gas on the 
insect and the tree was almost simultaneous. Several of these trees 
were so severely injured that they died the following season. On 
another occasion an 8-year-old apple tree not infested with borers or 
injured in any way was treated with an ounce of p-dichlorobenzene. 
